SR很高, 但PI, PO为0, 应该如何调整参数?
大家好,
请教一个问题,vmstat显示换页扫描很频繁,但pi po都不高,应该如何调vmo...谢谢。
System configuration: lcpu=8 mem=11840MB
kthr memory page faults cpu
----- ----------- ------------------------ ------------ -----------
r b avm fre re pi po fr sr cy in sy cs us sy id wa
3 7 1403172 5116 0 0 0 77692 270239 0 4541 14419 3337 11 33 20 36
3 6 1403174 5073 0 0 0 77887 272485 0 4561 13041 3126 11 33 19 37
2 7 1403174 4992 0 0 0 76983 283100 0 4491 13896 3203 12 32 21 35
2 8 1403173 5105 0 0 0 73835 238342 0 4290 12596 3163 11 31 21 36
2 6 1403174 5248 0 0 0 76705 305434 0 4515 13283 3187 12 33 21 35
vmo -a
cpu_scale_memp = 8
data_stagger_interval = 161
defps = 1
force_relalias_lite = 0
framesets = 2
htabscale = n/a
kernel_heap_psize = 4096
kernel_psize = 4096
large_page_heap_size = 0
lgpg_regions = 0
lgpg_size = 0
low_ps_handling = 1
lru_file_repage = 0
lru_poll_interval = 10
lrubucket = 131072
maxclient% = 90
maxfree = 1088
maxperm = 2617963
maxperm% = 90
maxpin = 2447184
maxpin% = 80
mbuf_heap_psize = 65536
memory_affinity = 1
memory_frames = 3031040
memplace_data = 2
memplace_mapped_file = 2
memplace_shm_anonymous = 2
memplace_shm_named = 2
memplace_stack = 2
memplace_text = 2
memplace_unmapped_file = 2
mempools = 2
minfree = 960
minperm = 145441
minperm% = 5
nokilluid = 0
npskill = 12288
npsrpgmax = 98304
npsrpgmin = 73728
npsscrubmax = 98304
npsscrubmin = 73728
npswarn = 49152
num_spec_dataseg = 0
numpsblks = 1572864
page_steal_method = 0
pagecoloring = n/a
pinnable_frames = 2418636
pta_balance_threshold = n/a
relalias_percentage = 0
rpgclean = 0
rpgcontrol = 2
scrub = 0
scrubclean = 0
soft_min_lgpgs_vmpool = 0
spec_dataseg_int = 512
strict_maxclient = 1
strict_maxperm = 0
v_pinshm = 0
vm_modlist_threshold = -1
vmm_fork_policy = 1
vmm_mpsize_support = 1
如果你对这篇内容有疑问,欢迎到本站社区发帖提问 参与讨论,获取更多帮助,或者扫码二维码加入 Web 技术交流群。
绑定邮箱获取回复消息
由于您还没有绑定你的真实邮箱,如果其他用户或者作者回复了您的评论,将不能在第一时间通知您!
发布评论
评论(9)
看看系统的lrub进程是否占用cpu比较高
是的,请问如何调参数能使lrud降下来。
有性能问题吗?没有不用关心
减少maxperm% = 90
以及maxclient% = 90
到70看看呢
或者增大maxfree 的值
最好能给个topas的页面或者nmon--m的页面看看
有性能问题呀,看vmstat的关于CPU的值就看得出来的,系统基本不在干正事,都耗在sys和wait上了。。。
还有一个问题就是io比较高,这会影响到SR吗?但page space是建在rootvg上,而rootvg的磁盘并不忙。
topas:
Wed May 18 01:37:04 2011 Interval: 2 Cswitch 4996 Readch 0.3G
Syscall 21371 Writech 28335
Kernel 29.8 |######### | Reads 2450 Rawin 0
User 13.5 |#### | Writes 11 Ttyout 376
Wait 39.5 |############ | Forks 0 Igets 0
Idle 17.2 |###### | Execs 0 Namei 26
Runqueue 2.0 Dirblk 0
Network KBPS I-Pack O-Pack KB-In KB-Out Waitqueue 13.5
en4 2.9 16.5 8.0 1.4 1.5
lo0 0.2 2.0 2.0 0.1 0.1 PAGING MEMORY
en5 0.0 0.5 0.0 0.0 0.0 Faults 1835 Real,MB 11840
Steals 67306 % Comp 52.8
Disk Busy% KBPS TPS KB-Read KB-Writ PgspIn 0 % Noncomp 47.9
hdisk2 0.0 0.0 0.0 0.0 0.0 PgspOut 0 % Client 47.9
hdisk0 0.0 14.0 2.5 12.0 2.0 PageIn 67123
hdisk1 0.0 6.0 1.5 4.0 2.0 PageOut 9 PAGING SPACE
cd0 0.0 0.0 0.0 0.0 0.0 Sios 67456 Size,MB 6144
hdisk16 0.0 11.3K 281.5 11.3K 0.0 % Used 1.0
hdisk18 0.0 10.9K 184.0 10.9K 18.0 NFS (calls/sec) % Free 98.9
hdisk8 0.0 10.8K 263.5 10.8K 0.0 ServerV2 0
hdisk15 0.0 10.4K 174.5 10.4K 0.0 ClientV2 0 Press:
hdisk5 0.0 11.1K 184.5 11.1K 4.0 ServerV3 0 "h" for help
hdisk7 0.0 11.2K 187.0 11.2K 0.0 ClientV3 0 "q" to quit
hdisk6 0.0 10.6K 266.5 10.6K 0.0
hdisk20 0.0 10.6K 176.0 10.6K 0.0
hdisk12 0.0 10.9K 266.0 10.9K 0.0
hdisk10 0.0 11.0K 265.5 11.0K 0.0
Name PID CPU% PgSp Owner
lrud 16392 8.2 0.6 root
oracle 2985996 3.0 6.2 oracle
oracle 2936914 2.9 6.2 oracle
oracle 3432840 2.8 5.0 oracle
oracle 2830548 2.3 5.0 oracle
oracle 3518916 2.0 6.1 oracle
oracle 3522938 1.9 5.0 oracle
oracle 3576262 1.7 6.2 oracle
oracle 3797408 1.6 5.1 oracle
oracle 2805888 1.3 5.0 oracle
oracle 3707334 1.3 5.0 oracle
oracle 3494160 1.2 5.0 oracle
oracle 3838436 1.1 6.0 oracle
oracle 3834356 1.1 5.0 oracle
oracle 2977840 1.0 5.0 oracle
oracle 3174464 0.7 5.0 oracle
oracle应用?
使用的是文件系统?!
在给个mount的结果看看呢